Two weeks after a Government Accountability Office (GAO) report warned of potential gaps in the GPS satellite constellation, reassurances from Department of Defense (DoD), U.S. Air Force, GPS program officials, and industry are slowly restoring calm to an anxious public.
Coupled with a May 7 congressional hearing (and subsequent media coverage) that fanned the flames, the report described scenarios — for instance a two-year setback in launching the first GPS III spacecraft — that could lead to deterioration in the quality of GPS service due to delays in building new generations of satellites and past program management problems. The news coverage drew primarily on the GAO report and prepared statements presented at the hearing and not on the much more interesting dialog with subcommittee members captured on video or on the assumptions and methodology underlying the constellation analysis.
Two weeks after a Government Accountability Office (GAO) report warned of potential gaps in the GPS satellite constellation, reassurances from Department of Defense (DoD), U.S. Air Force, GPS program officials, and industry are slowly restoring calm to an anxious public.
Coupled with a May 7 congressional hearing (and subsequent media coverage) that fanned the flames, the report described scenarios — for instance a two-year setback in launching the first GPS III spacecraft — that could lead to deterioration in the quality of GPS service due to delays in building new generations of satellites and past program management problems. The news coverage drew primarily on the GAO report and prepared statements presented at the hearing and not on the much more interesting dialog with subcommittee members captured on video or on the assumptions and methodology underlying the constellation analysis.
In fact, the long-known issue of constellation sustainment probably masked the equally substantive matter of “synchronizing” modernization of the GPS space and ground segments with development of military user equipment, and GAO’s recommendation for a “single authority to oversee the development of the GPS system.”
After all, the DoD’s biennial report to Congress about GPS, submitted last October had (once again) pointed out the condition of the constellation, “Of the 31 GPS satellites currently on orbit, 20 are past their design life, and 19 are without redundancy in either the navigation mission equipment or the satellite bus, or both.”
And DoD didn’t have to look as far ahead as GPS III for capability risk scenarios: “Should GPS IIF launches be delayed, sustainment of the GPS constellation will be difficult, and the [U.S. government] could fail to meet performance levels prescribed in published federal plans and standards.”
Nonetheless, the high-profile GAO report clearly resonated more with both intended and unintended audiences, gaining extra mileage at a May 7 hearing of the House Committee on Oversight and Government Reform, subcommittee on National Security and Foreign Affairs. The subcommittee, chaired by John Tierney (D-Massachusetts) had requested in 2007 that the study be undertaken by the GAO, a Congressional watchdog agency.
Cristina Chaplain, director of acquisition and sourcing management, GAO, talking about objectives of the GPS report.
Responding to a question from Tierney, Cristina Chaplain, the report’s primary author and the GAO’s director of acquisition and sourcing management, said, “What we were hoping to see was a strengthened leadership focus on GPS because of the potential capability gaps, because of the risks in acquisition, and because of the criticality of GPS to everybody in the nation.”
She got that and more. But at the end of the day (and at the end of this article), Chaplain wonders whether the report will make any substantive difference.
Much Ado about Something
In the days following release of the report and Tierney’s hearing, the airwaves and Internet were literally atwitter with reports, rumors, and reactions to the GAO report and hearing. Government agencies began fielding calls and e-mails about an impending doom for the GPS system.
GPS receiver manufacturers and electronics retailers heard from worried customers who were considering delaying purchases of receivers, incited by headlines such as PC Week’s May 20 query, “With a GPS Failure Possible, Is It Still Safe to Buy?”
As the turmoil mounted, the Air Force Space Command (AFSPC) at Peterson AFB, Colorado, held a May 20 “tweet forum” on the subject, moderated by Col. Dave Buckman, AFSPC’s command lead for positioning, navigation and timing.
Given the constraints of Twitter messaging, the Internet news conference produced more novelty than news, but insisted that GPS was not going down. Concluded one tweeter, “The issue is not whether GPS will stop working. There’s only a small risk we will not continue to exceed our performance standard.”
AFSPC followed up the next day with a press release in which it noted that the command had “acknowledged the potential for an availability gap years ago” and set out to ameliorate the situation. “I have high confidence we will continue to sustain at least the 24 satellites required to maintain our current performance standard,” the release quoted Gen. C. Robert Kehler, AFSPC’s commander.
“The Air Force has been a good GPS steward continually providing ‘better than expected’ service to our GPS users. At this point, we foresee no significant loss of service in the future, near or far,” Kehler added.
The Boeing Corporation, prime contractor for the IIF, weighed in with an update on its new satellites. “Working very closely with the Air Force and its team,” the company said in a press release, “Boeing has taken aggressive steps to resolve the technical issues on IIF with a strong emphasis on Mission Assurance.”
Design changes required to ensure performance over the satellite design life caused schedule delays, Boeing said, adding that “these changes are in the final phase of implementation, and a fully integrated satellite (SV1) has already successfully completed the thermal-vacuum test program — the most stressing system level test.”
In an Air Force/Boeing effort to expedite launch of the first IIF, now expected early next year, the second satellite was shipped to the Cape on May 6 to perform system-level compatibility tests and “serve as a risk-reduction pathfinder for SV1 processing later this year.”
Some of the more wild-eyed press reports speculated the GPS sky was literally falling, or at least satellites were falling out of it.
Fox News headlined its story, “GPS System Could Fail Next Year,” and CBS News warned, “GPS On Verge Of Break Down.” The United Kingdom’s Express worried that “Motorists face Chaos over GPS ‘Meltdown’’ and (a personal favorite) from Good Morning Silicon Valley, “The GPS says I’m in your driveway, but the car’s sinking and all I see is cattails.”
The latter blog post began, “If the Air Force and its contractors don’t get their act together pretty quickly, in a couple of years your car’s navigation system may be giving you instructions like, ‘In a mile or so, turn right,’ or ‘You have reached your destination, more or less.’”
(Inside GNSS’s contribution to the furor, “GAO Says GPS III Launch Delay Could Drop Constellation below 24 Satellites,” underlined the agency’s most dire scenario for the 2016–19 timeframe, but quickly pointed out that the GPS IIIA program is on schedule and followed a few days later with an update on the program’s continuing progress.)
Satellite Gap or Credibility Gap?
Much of the heat came from simulations or modeling of potential declines in the number of satellites in the constellation. For example, the GAO suggests that a two-year delay in the production and launch of the first and all subsequent GPS III satellites “would reduce the probability of maintaining a 24-satellite constellation to about 10 percent by around fiscal year 2018.”
However, with 30 GPS satellites now operating on orbit, another Block IIR scheduled for launch in August, and 12 more IIFs coming on-line beginning next year, a lot stands in the way of that happening. And whatever happens will probably take place slowly, with a gradual decline in positioning accuracy and availability in some locations due to fewer satellite signals being received.
Steve Huybrechts, principal director for command, control, communications, space, and spectrum in the office of the assistant secretary of defense for network and information integration (ASD/NII), explaining “singular responsibility” of his office for GPS.
“We’re using the term gap and that sounds pretty black and white, but the GPS constellation degrades more gracefully because of the number of satellites,” Steve Huybrechts, principal director for command, control, communications, space, and spectrum in the office of the assistant secretary of defense for network and information integration (ASD/NII), told the subcommittee.
“We shouldn’t be sitting here thinking that all the GPS receivers are going to stop working,” Huybrechts said. “What you’re going to get is a slight degradation in performance over small portions of the world over short periods of time compared to today.”
Moreover, the scenarios — Monte Carlo simulations — were not really new, they merely repeated those conducted for many years on behalf of the Air Force. The analysis was developed and is overseen by the Aerospace Corporation, a federally funded research and development center (FFRDC) with headquarters in El Segundo, California, that provides independent technical and scientific research to the GPS Wing among others.
“We basically mirrored what Aerospace does and did not invent a new methodology,” Chaplain told Inside GNSS.
In response to questioning at the subcommittee hearing, Maj. Gen. William McCasland, director of space acquisition for the Office of the Undersecretary of the Air Force, acknowledged that the GAO had followed “the methodology and technical assumptions used by the Air Force” to predict future performance of the constellation, noting that they had given those details to GAO a year ago.
“In that time, some things have changed,” McCasland continued. “These lifetime assumptions are a bit like actuarial tables with people, except that we don’t have human history to base them on. We’ve got a much shorter history. The specific population we base it on is the flying population of GPS IIR satellites, and in this year, the satellites that we base this model on — their failure forecast — have grown a little bit. So, just in the past year we can look to this gap and, if we were to recalculate it, it would be about half the depth that it is today.”
Chaplain later said that, in fact, the GAO had held up its report to repeat the modeling exercise in March 2009 and, as a result, considered it a current evaluation of the constellation’s prospects.
Behind the various scenarios, including a slight (10 to 20 percent chance) of a sub–24 SV constellation beginning next year, lies a complex and evolving exercise in prediction called the GPS Generalized Availability Program (GAP). The GAP seeks to calculate the reliability expectations of on-orbit and unlaunched space vehicles (SVs) and help the Air Force match those expectations with development, manufacturing, and launch of new GPS satellites.
McCasland said that he believes the Air Force “has the gold standard” of satellite prediction models.
For nearly 20 years, the process has been managed at Aerospace by Jim Womack, who holds a Ph.D. in statistics from the University of California at Santa Barbara.
The GAP analysis evaluates and tracks the performance of subsystems and critical components on the satellites. Of particular importance are the satellite’s reaction wheels, atomic clocks, and solar panels — all items crucial for the spacecraft’s operation and all of which eventually wear out. A fourth factor, random electronic failures, is also modeled in the analysis.
In predicting the longevity of the various satellites and their subsystems, the GAP analysis considers different factors at various points in a satellite’s development and operation. In a satellite’s initial design phase, reliability models are usually based on lifetime specifications. As subsystems designs mature, detailed “piece part” models are developed based on failure rates specified in a DoD handbook MIL HDBK 217.
The final stage of reliability modeling is the incorporation of subsystem failure models that use on-orbit failure data from operating satellites.
Presumably, the more experience that the Air Force and Aerospace Corporation has with satellites on orbit, the more refined their analysis becomes. However, the Block IIRs that began being launched nearly 12 years ago have experienced so few component failures that modeling their future life probably includes a larger unknown element.
One thing to which the model doesn’t have access, according to a GPS satellite expert, is an extensive, accelerated long-life “stress test” program of spacecraft subsystems and components on the ground, which could further reduce the uncertainty of the wear-out rates, especially of new devices.
Other Factors
The Air Force reports regularly on the current status of the constellation, enumerating the subsystem failures of the on-orbit satellites. As of last September all of the clocks and reaction wheels of the oldest Block IIR satellite — SVN43/PRN13 launched in July 1997 — were operational.
In addition to these factors, McCasland said that the model is also based on a predicted launch rate and a full use of all the power on the satellite for all the payloads, such as the Nuclear Detonation (NUDET) Detection System (NDS). (In September 1997, for instance, the launch and production schedule for Block IIRs and IIFs expected the last IIR to be launched in July 2002 and the first IIF that October.)
“So, the operator will have choices to make, about how fast they launch satellites and how they spend the power on the satellite,” McCasland said, adding “We’re confident that we’ve got several degrees of margins in preventing a gap like has been presented in the GAO report.”
Huybrechts later noted that NDS “requires a lot fewer” than 24 satellites to operate but that the payload had been launched on all satellites in a block to enable manufacturers to build to a single SV specification.
Moreover, in addition to the 31 operating satellites now on orbit — including SVN49, launched March 24 and still in the on-orbit check-out phase — three GPS “residual satellites” remain in orbit. These are older spacecraft with “suspect” subsystems but which can still be brought back on-line.
These satellites are brought out of the residual mode and tested every six months, according to Lt. Gen. Larry James, commander of the 14th Air Force (Air Forces Strategic), Air Force Space Command and of the Joint Functional Component Command for Space, U.S. Strategic Command.
“The GPS constellation isn’t an on-or-off thing,” James told the subcommittee members, adding as an example, “We can adjust the constellation to put less accuracy over a less populated area.”
Finally, the Air Force and DoD testimony emphasized the officials’ belief that, despite the history on past GPS satellite programs, the current GPS IIIA program will deliver on time for a first launch in 2014.
The prime contractor, Lockheed Martin, and Air Force engineers recently completed a preliminary design review for the program, which remains on track one year into a six-year schedule.
In a recent press statement on GPS III, Gen. Tom Sheridan, commander of the Space and Missile Systems Center at Los Angeles AFB, said, “New acquisition approaches, including phased acquisition and prototyping, will reduce risk to constellation sustainment in the future.”
And the GPS Wing commander in charge of GPS acquisition, Col. David Madden, received a glowing endorsement from Huybrechts during the congressional hearing. “Col. Dave Madden, who I have the highest respect for, . . . is one of the better program managers that I’ve met.”
Getting on the Same Page
Despite the reassurances that the acquisition process has gotten “back to basics” and that sustaining the GPS constellation is receiving the attention it needs, the other issue highlighted in the GAO report remains: the dispersed authority for the GPS program overall.
Nominally, ASD/NII has overall responsibility for GPS at the Office of the Secretary of Defense (OSD), and the Air Force Space Command takes care of system operations. The Space-Based PNT Executive Committee is supposed to coordinate interagency activities. The GPS Wing establishes the specifications for space and ground infrastructure and military basic receiver technology, but individual services are responsible for modifying GPS equipment for specific platforms.
GAO said that this situation needed to change, that development of the space and ground segments needed to be better coordinated with development and fielding of user equipment.
Huybrechts acknowledged that “about four to five years ago, the [DoD], and especially the Air Force, recognized that there was a risk of a [satellite] gap if we did not act. For that reason, with the [funding] reserves available to the program, we prioritized the space segment, followed by the ground segment upgrades, higher than the user equipment.”
DoD and Air Force officials insisted that this situation — what Tierney characterized as taking from Peter to pay Paul — has been reversed.
At one point in the hearing Tierney asked Huybrechts directly, “Are you the one drawing everything together and making sure that they are coordinating and getting things done in a timely fashion?”
“That is my role up at the OSD level,” Huybrechts replied. “And we have put a single service, the Air force, in charge of all aspects of the GPS program. From that perspective you do have a single entity in charge of acquisition and operation of the system.”
This is unlike the way DoD handles many of its other space programs, where many services are involved, according to Huybrechts.
“My office at the OSD level has been given singular responsibility by the deputy secretary of defense for this program,” he continued.
“That said, we have to manage the program within the department’s processes. It’s one of many programs and has to be traded off against the department’s various program needs.”
Which wasn’t exactly what GAO was hoping for.
“In its comments, DoD did not indicate any big changes would be made with the leadership structure, though it concurred with our recommendation,” Chaplain told Inside GNSS. “Ultimately, we were hoping for a change that would result in high-level, focused attention — high enough to ensure resources are targeted in the right places and potential problems get quick attention. Especially when it comes to the user assets because the current structure has not resulted in effective synchronization.”
Although GAO did not state it explicitly in the report to Congress, Chaplain said the agency was thinking of a solution in terms of a” GPS czar.” Hopefully, with more attention on GPS, this will happen by default,” she concluded.
